coulamb cutoff , chi2 value

[sohan]

Dear sir,
The attached are /chi2vsenergy plot for a material (less peaks- ipa , other -TDBSE) ,
q.1. I wanted to know if these 2 are valid ?
q.2. Is it neccessry to use supercell approach for a 2d system ? (i didn't use supercell in the attached )
q.3. how we calculate a certain /chi2 value for a system that defines the system , does the chi2_magnitude correspnd to the peak is that value?
q.4. what is the role of coulamb cutoff (-r) , how to take the appropriate value of it ?

best
sohan

[claudio]

Dear sohanq

>1. I wanted to know if these 2 are valid ?

yes, in principle both are valid if calculation are well done

> q.2. Is it neccessry to use supercell approach for a 2d system ? (i didn't use supercell in the attached )

yes, you should use a super-cell large enough to accommodate the cutoff
https://wiki.yambo-code.eu/wiki/index.p ... al_systems

>q.3. how we calculate a certain /chi2 value for a system that defines the system , does the chi2_magnitude correspnd to the peak is that value?

this question is not clear to me, the chi2 that you calculate correspond to the chi2 of the cell you used, in case of 2D system
you have to re-scale it to the effective thickness of the system as described here:
https://wiki.yambo-code.eu/wiki/index.p ... :_2D_or_1D

>q.4. what is the role of coulamb cutoff (-r) , how to take the appropriate value of it ?

the cutoff is used in the calculation of the collisions and TD-BSE. Regarding the appropriate value
if you choose "slab z" you do not need to set any other value.
CUTGeo= "slab z"
RandQpts=3000000 # [RIM] Number of random q-points in the BZ
RandGvec= 1 RL # [RIM] Coulomb interaction RS components

best regards
Claudio

[sohan]

Dear sir,
Thank you for prompt reply. The chi2 value that you mentioned for hBN in this paper https://doi.org/10.1103/PhysRevB.89.081102 table 1 . Is this chi2 value at zero frequency ?
and supercell is being used to make it 3d periodic system as yambo requires ? sir the spectra i obtained is for 2d system but didn't used supercell , so it shouldn't be correct?
best ,
sohan

[Daniele Varsano]

Dear sohan,

please sign your post with your name and affiliation, this is a rule of the forum. You can do once for all by filling the user profile.

2D system in Yambo, as well as all plane wave methods, are treated in a supercell approach.
This consists in adding vacuum in the non-periodic direction to eliminate spurious interactions. On top of this, you can use coulomb cuotff technique to speed up the
convergence with respect to the amount of vacuum needed to isolate the system.

It is not clear what you write:
1) supercell is being used to make it 3d periodic system as yambo requires ?

All the system in plane wave are periodic, supercell is used to isolate the system

2) the spectra i obtained is for 2d system but didn't used supercell

What do you exactly mean with "I did not used supercell?).

Best,

Daniele

[sohan]

Dear sir,
thank you for the prompt reply. sir yambo treats only 3d system , so in order to make it work with 2d sytem we requires its supercell. but in attached I took only the unit cell of a system and tried producing the SHG, so my concern was if these 2 curve are said to be wrong or not? Also in the example https://yambo-code.eu/wiki/index.php/Se ... _of_2D-hBN , if we look at DFT inputs -- unit cell of hBN is been taken. Or is there any other way around to take supercell ?
best
sohan

[Daniele Varsano]

Dear Sohan,

if we look at DFT inputs -- unit cell of hBN is been taken

the example you mention is a monolayer of hBN. If you look at the DFT inputs this is not the unit cell of bulk hBN, but vacuum has been considered in the direction perpendicular to the hBN sheet:

&system
ibrav = 4,
celldm(1) = 4.716
celldm(3) = 7.
..
&

pay attention to celldm(3) value.

Best,

Daniele

[sohan]

Dear sir,
Thank you for the valuable insights. i have one query regarding NLbands , sir i have read in a paper where MoS2's NLbands were taken b/w 19-34 and total bands were 300 bands.
Do we get the info abt NLbands range (consisting of 2 valance and 2< conduction bands near fermi ) from bands.out (attached) where band indexing is being done ?
and can we keep the same range for TDBSE - shg as well?
Also how to get /chi2 value at /omega(freq.) = 0 ? as setting ENrange from 0ev to 8ev give no plot.

[Daniele Varsano]

Dear Sohan,

you have the info on band indexing in the r_setup file. There it is also reported which bands are VBM (maximum valence band) and MCB (minimum conduction band). From there, you can select the bands entering in the spectrum. I suggest you to start with few bands across Fermi and then increase the range up to convergence.

Best,
Daniele

[lorenzo.sponza]

Dear Yambo users and developers,
I post my question inside this thread, but feel free to move it if you prefer.

I am running an IP calculation with Yambo 5.1 on a bilayer of hBN. Among my input keys I set

Code: Select all

RandQpts=5000000         
RandGvec= 10   RL
CUTGeo= "slab z" 

The calculation runs correctly with no particular issue, and results seems consistent. Though, I am perplexed about the right interpretation of some lines of the report, namely:

Code: Select all

[RD./test_reading_databases_bse_db//ndb.RIM]------------------------------------
Brillouin Zone Q/K grids (IBZ/BZ)                :   15   81   15   81
*WRN* Coulomb cutoff potential                   : none
Coulombian RL components                         :  11
Coulombian diagonal components                   : yes
RIM random points                                :   5000000

and few lines later

Code: Select all

[RD./test_reading_databases_bse_db//ndb.cutoff]---------------------------------
  Brillouin Zone Q/K grids (IBZ/BZ)                :   15   81   15   81
  CutOff Geometry                                  : slab z
  Coulomb cutoff potential                         : slab z
  Box sides length            [au]                 :  0.000000  0.000000  0.000000
  Sphere/Cylinder radius      [au]                 :  0.000000
  Cylinder length             [au]                 :  0.000000
  RL components                                    :   57263
  RL components used in the sum                    :   57263
  RIM corrections included                         : no
  RIM RL components                                : 0
  RIM random points                                : 0

Q1: Why there's a warning in the first part as if there were no Coulomb truncation? Isn't "slab z" meant to do precisely this?
Q2: Why RIM components and points vanish in the second part of the report?
Q3: Am I actually setting the calculation properly?

Many thanks in advance.

[Daniele Varsano]

Ciao Lorenzo,
it is possible there is some inconsistency between what is calculated and what is reported.
In order to inspect it, it would be desirable if you can repeat the calculation using the latest release of Yambo (5.3) and attach your entire report if the inconsistency persists.

Best,

Daniele